Mesa/progs/tests/seccolor.c gives bad results for tcl_mode=1 on Radeon. I
haven't figured out why.
Same goes for r200. Probably the same issue for both.
Created attachment 3659 [details] [review]
patch to fix secondary color issue when lighting is disabled on radeon and r200
suggested fix. We may need to transmit the secondary color values even when
lighting is disabled (in particular, when color sum is enabled). (I don't have
time right now to test/commit, feel free to do it.)
Yeah, that's right, good catch. (I'd been looking in UpdateSpecular for what
was wrong, instead of there). Tested on rv200 and r200 with seccolor, and
Mass version move, cvs -> git